Wednesday, 12 March 2014

Big data startup Sociocast goes beyond analytics with new SaaS tools (exclusive)

yamandiuka

Big data and analysis startup Sociocast has launched two new Software-as-a-Service tools aimed at helping advertising and media companies better understand their customers, the company revealed today.
Sociocast’s software analyzes “big data” sets and deliverspredictive real-time audience data to help companies make more informed marketing decisions. Sociocast CEO Albert Azout told VentureBeat that it can’t disclose specific customers yet, but said it works with many large data aggregators, ad networks, and agency trading desks.
The company’s significantly updated software tools are versions 2.2 of Sociocast Connect and Sociocast Signal, both of which advance the company’s mission of creating “actionable” intelligence.
“Unlike other companies, we’re not just creating intelligence and analytics,” Azout told us. “We’re also working with decision makers that are actually buying media. That means you get better and more meaningful data on the fly.”
Sociocast Connect, the company’s flagship product, now offers the following features:
• Data Transformation: Sociocast Connect translates brands, ad networks, and large data owners’ data into highly segmented, comprehensive user profiles and audience clusters using first and third-party data sources. These highly accurate data outputs can plug directly into clients’ decision engines.
• Enhanced Real-Time Analytics: The new platform can provide analysis on more than 2 million time-stamped interactions per minute. What’s more, this information will identify contextual attributes that serve as the basis of a brand or advertiser’s “signature”.
• Private Cloud: Sociocast Connect operates a multi-tenant framework that isolates clients’ data in order to ensure enhanced security. The platform also includes cutting-edge security features that prevent data leakages.
• Easy Integration: Sociocast Connect supports dynamic sFTP batch data transfers and seamlessly complements 3rd-party platforms including DMPs, SSPs, and DSPs.
And Sociocast Signal now offers the following:
• Mobile Detection Services: Sociocast will process mobile and app logs with greater speed and accuracy, as well as increased coverage of mobile sites for classification.
• Coverage & Granularity: The new platform introduces an extremely logical classification taxonomy, as well as classifications into 1 million different people, places, things, and brands.
Sociocast’s data-crunching acrobatics are run out of a data center with 100 servers in Virginia. Interestingly, Azout said the company had looked at Amazon’s highly touted public cloud as an option, but it was too expensive and had too much latency for Sociocast’s needs.
New York-based Sociocast has raised about $5 million since it launched in early 2010. Its latest $1.5 million round was led by Detroit Venture Partners and angel investors Bobby Yazdani, Joe Zawadzki, and Doug Imbruce.

For Earth Day: a chat with IBM’s “Building Whisperer,” Dave Bartlett, on the tech behind sustainability

yamandiuka

IBM recently won the contract to create a more sustainable and energy efficient Los Angeles Unified School District. But before ripping out old roofs or replacing leaky windows, David Bartlett’s, the computing gaint’s sustainability guru, took a look at the data.
“Most buildings don’t have a good way to look at all their systems holistically,” Bartlett told me. “It’s like buying a car without a dashboard to tell you when the oil is running low or the engine is running too hot.” Buildings account for 42 percent of the worlds energy use and emit more carbon dioxide than cars.
With the availability of cheap, wireless sensors, it’s now possible to measure nearly every facet of a building’s energy use. And what Barlett found in L.A. was striking. “You had one air unit that was blowing full on hot air, and another that was blowing full on cold air, working against each other, in the same building.”
Data, data everywhere, now let’s stop and think. That’s the big data model that Bartlett is hoping IBM can impart. “For a public company, an upfront investment in energy improvement can be a hard sell. It may have a long term benefits, but they answer to the market every quarter.”
Bartlett’s message is simple. “We’re not asking you to rebuild anything. Let’s look at what you’re working with and see how we can save money through efficiency. We are no longer held back by tech. Most buildings are drowning in data.” Energy typically accounts for 30 percent of a buildings annual costs, so savings through sustainability can have a big impact on a company’s bottom line.
The average consumer might relate to Bartlett’s work by thinking of a product like the Nest learning thermostat. IBM is working with big corporate clients, but the same principles apply. Visualizing data, layering on machine learning that tracks user behavior, and making an intuitive interface that non-technical experts can grasp when dealing with heating or cooling their homes or offices.
Bartlett specializes in what he calls miniature cities, like Tulan University, which is being completely rebuilt after the flooding of Hurricane Katrina, is working with Bartlett to create smart systems from the beginning that will lay a foundation for a more efficient, environmentally friendly campus.
But computers are only part of the story. “We always looking for ways to get people involved,” says Bartlett. At the LA School Dsitrict, IBM created a smartphone app, in partnership with the startupCitysourced, that lets students and teachers photograph, geotag and report things like cracked windows or broken air conditioners. “Crowdsourcing is a powerful way to augment our data. I always say, when it comes to new gathering information, humans are the smartest sensors.”

MetaMarkets crunches your big numbers, raises $15M

yamandiuka

Big data analytics company MetaMarkets has raised $15 million in a new funding round, the company announced today.
MetaMarkets originally analyzed prices for online ads on mobile and regular websites for media companies in real-time. Now the company has expanded its reach to offer big data analytics for social and e-commerce companies that need to crunch a lot of numbers. It’s data science-as-a-servicehelps analytics teams notice trends and forecast how future events will affect their data.
In this increasingly crowded big data startup market, MetaMarket’s competitors are any one of the big data companies, such as ConnotateInsightsOne, or OpenView. Each big data company is looking to find its niche in the market, but they are all taking huge amounts of unstructured data and trying to make sense of it.
Khosla Ventures led the round, which will used to push the company forward in its new focus. Existing investors IA Ventures and True Ventures also participated.
MetaMarkets is based in San Francisco, Calif. and has raised a total of $23 million.

Genome entrepreneurs say their data will help you live longer

yamandiuka

The cost of sequencing the human genome continues to fall, reaching a low of $1,000 this year due to a new microchip and machine designed by genetics company Life Technologies Corp. And unleashed by those lower costs, a small cadre of entrepreneurs in Silicon Valley is exploring ways to harness this data to enable us to live longer and healthier lives.
Dr. Dietrich Stephan, a human geneticist, has spent the better part of a decade trying to achieve that goal. Until recently, it has been costly and time-consuming to map the 3 billion units of DNA, known as base-pairs, which make up the human genetic code. But now, he said, with the low cost of gene sequencing technologies, we are on the brink of banishing a one-size-fits-all approach to medicine.
“Every disease has a genetic component, and yet largely none of the available genetic information is being used today to treat patients,” he said.
Genome sequencing had been limited to a select few. Biographer Walter Isaacson claims the late Steve Jobs had his DNA sequenced for $100,000 after being diagnosed with pancreatic cancer.
To biotech entrepreneurs, bringing gene sequencing to a mass market is the most exciting development since the completion of The Human Genome Project. According to Joe Betts-Lecroix, a biophysicist and entrepreneur, that major international undertaking was expected to yield a map of the mutations that cause disease. Instead, genetics researchers found that the genes only seem to account for a tiny percentage of inherited traits and differences between a healthy and sick person.
“The Human Genome Project was a monumental achievement, but also a huge disappointment,” said Betts-Lacroix, who recently delivered a rousing TEDx SF talk on using genetics to cure aging.
“It’s not about the human genome, but more the human gene pool. To really understand what’s going on, we’ve found that we have to sequence the DNA of millions of people. This work is only beginning, and will require much lower-cost sequencing. Fortunately, that seems to be upon us,” said Betts-Lacroix.
At a select few hospitals across the country, the work has already begun. In 2010, the Children’s Hospital in Boston recruited Dr. Dietrich Stephan – a biotech entrepreneur fresh from the success of Sequoia-funded genetics startup Navigenics – to join the Gene Partnership Initiative, one of the first efforts to gather genetic information about disease. Relying on his experience as a biotech entrepreneur, Stephan developed the hospital’s first sustainable program to sequence the genomes of 100,000 young patients.
The program is still ongoing, and researchers are finding links between genes, the environment, and complex genetic disease. Already, the flood of genetic information is being used to produce drugs that are better targeted to patients. Not everyone processes medicine in the same way; by understanding patients’ genomic make-up, doctors can provide a therapy they’ll respond to best. That’s an important capability considering that this year alone, close to 800,000 people were injured or died due to adverse reactions to drugs.
“With the knowledge we have now about your genes, we are finding ways to diagnose sick people earlier and develop medicines to zap the core defect,” said Stephan of the Gene Partnership Initiative.
Biotech entrepreneurs predict that the low-cost of gene sequencing technologies will yield new approaches to the treatment of cancer and other diseases by 2015. With this in mind, Stephan has delved into clinical diagnostics with a new startup, Silicon Valley Biosystems. With his latest venture, he said the goal is to “get fancy IT into the hands of physicians so patients can benefit from it.”
“It’s no longer business as usual in medicine,” he said. Patients will be “touched and informed by genetics” in the next 2-5 years, meaning that hospitals will increasingly adopt gene-sequencing technologies to better treat and diagnose disease.
Sabah Oney, business development lead at SVBio, said we’ll see common diseases being re-branded as we better understand their root cause. “Instead of diagnosing breast or lung cancer, physicians will refer to the disease by its genetic abnormality.”
The major challenge facing genetics entrepreneurs is to take the concept from the lab and turn it into a commercial product. “The rate of failure discourages many investors,” said Mohsen Moazami, founder and general partner of Seif Capital, an early stage venture capital firm focused on health and bio-sciences. “But there’s no question in my mind that for a subset of companies that crack the code on this effort, the opportunity is immense,” he added.
The startup that firmly put the spotlight on genetics is 23andme, which provides insights on your future and ancestral past based on a DNA sample. This month, the company hit a new milestone: 150,000 users. 23andme does not sequence the entire genome. Instead, it determines through genotyping whether users are at risk of developing particular diseases.
Genomic sequencing is the future, but 23andme boasts a vast database of genetic information. “This data is driving a whole new wave of discovery in terms of genes and environment. The real meaning will become clear in the next few years,” said Dr. Uta Francke, senior medical researcher for 23andme.
This year, 23andme has participated in research to understand how the body will react to certain medications. Francke recalls a recent study of how a female user in Australia discovered through the test that she was highly averse to anesthesia. “She may not have woken up. Some of these people say 23andme has saved their life,” said Francke.
“We think there is evidence that when people know their genes they take their health more seriously,” Francke added, although she did not point to supporting research. She suggests that users should print out their information and take it to their personal physician for advice about taking actionable steps to improve their health.
Critics of 23andme say that when this information is placed directly into the hands of consumers, very few choose to adjust their lifestyle accordingly. It took several years for 23andme to attract 100,000 users.
WellnessFX, a San Francisco-based health and wellness startup, is among the first to leverage human health data to prevent the onset of disease. The company’s target market are people who want to remain in peak physical health for longer. These users can access a visual spectrum of both genetic and biological data. The idea is that data is only meaningful when it’s used to make recommendations to improve general health. Users are re-tested at regular intervals to chart improvement.
“We show users that if they change their lifestyle and nutrition, they will see measurable results in a finite period,” said Jim Kean, chief executive officer of WellnessFX. Individuals can log in to explore a colorful display of their data at any time; Kean calls it a “personalized health dashboard.”
Innovation on the genetics data front has moved at a glacial pace and has yet to hit a mainstream market. But startups like SVBio, KnomeIllumina, and Navigenics have succeeded in raising awareness about how data can improve public health.
This March, the Obama administration showed its support with a $200 million grant for industries that can benefit from “Big Data.” With the increased availability of human data, it’s no surprise that the National Institute of Health (NIH) received the lion’s share. As part of this initiative, the world’s largest set of data on human genetic variation was uploaded on Amazon Web Services cloud (AWS) and is available to all researchers.
Betts-Lacroix, a serial entrepreneur working with startups like Halcyon Molecular to develop advanced DNA-sequencing technologies, said this open data will precipitate a gold rush for genetics research. As he puts it: “There are discoveries just waiting for people to find them, no lab work required.”

How to turn big data into engaging infographics with a single app

yamandiuka

Digesting big data can be migraine-inducing. Knoema wants to make it as easy as scanning a graph.
“With data, we want to do what YouTube did for video,” Vladimir Bougay, founder and CTO of startup Knoema, told VentureBeat in a recent interview. Knoema’s site lets users create their own simple infographics containing tables, charts, maps, images, and text to help them tell a story.
The process is DIY, but the end result is a slick-looking and shareable page that lets you select and display the patterns and trends you find in the data you or anyone else has uploaded. (We’ve included some sample Knoema user-created graphics below.)
That’s the other fun part about Knowma — the data itself. The startup has already collected more than 500 datasets on various topics and slapped a search engine on top to make navigation easier. You can also upload any data you have on hand. You could, if you so chose, mash up your own private data with data from public sources.
“I think it’s one of those businesses that would have increasing returns as people use it,” said Will Price, chief executive at Flite, after watching the company show off its site at the DEMO conference in Santa Clara, Calif. today. “I thought it was a very impressive product.”
Above all, said Bougay, the process is fast. “Users can build tables/charts very quickly, download data in Excel format, save data views, and share them with friends. It takes minutes, not hours,” the founder told us. “If you have a story based on data to tell, you will find all the necessary tools at Knoema.”
You can share data dashboards, complete with links, maps, charts, and graph with other users through the usual variety of social networks. You can also export them to PowerPoint or embed them on blogs or other sites.
But you know who really loves data? Educators. “We have exposed Knoema to a few universities in the U.S.,” said Bougay, “and a number of professors have decided to use Knoema as a platform for teaching.” Columbia Business School, for example, has taken advantage of Knoema for its healthcare business instruction.
Despite the attraction, there are concerns about its business model. “The go-to-market thing might be a challenge and will you have a credibility issue with the data. You know, how we criticize Wikipedia,” said John Dillon, chief executive of Engine Yard.
Bougay points out that open data and big data have been hot topics of late. “However, there is no one common understanding of what all these data issues are about and what does it practically mean for data users,” he said. “While data is all around the web, one would have to go through a number of tedious tasks to put together and share/disseminate a simple data-driven content … There are two practical problems that Knoema is meant to solve: finding relevant data and getting it in a ready-to-use format for content building.”
One of the bigger problems around exploring data is that normal search engines aren’t great at parsing requests for raw data or returning data-focused results, especially results you can immediately use or incorporated into research and story-telling. “The ultimate goal of using data is to make an argument, tell a story, or come up with an informative decision,” said Bougay.
The startup is currently working on adding more visualization options and analytics tools. Knoema is also working on a “simpler and smarter” UI with collaboration features for friends or teams to build dashboards and graphics together.
And, of course, the startup has an eye on developers. Bougay said he wants to make Knoema not just a consumer-facing product but also a platform for data-driven apps from third-party developers.
Knoema was founded in 2011 and is based in Washington, DC, with R&D and data teams located in Perm, Russia and Bangalore, India. Knoema has been bootstrapped to date and is looking for investors that share its vision.
[vb_gallery id=417380]
Knoema is one of 80 companies chosen by VentureBeat to launch at the DEMO Spring 2012 event taking place this week in Silicon Valley. After we make our selections, the chosen companies pay a fee to present. Our coverage of them remains objective.

Pipe dreams: ABB invests in water monitoring startup TaKaDu

yamandiuka

What would happen to your city if all the taps ran dry? We can’t survive without water, but water shortages will effect many of us in the future due to population growth, climate change and crumbling water networks.
Electricity infrastructure provider ABB, often described as one of the “Four horsemen of the power grid“, just boosted its water business by leading a $6 million investment in water infrastructure monitoring startup TaKaDu.
“Water is a key focus area for ABB and this investment will further strengthen our power and automation offering for the sector,” said Franz-Josef Mengede, head of ABB’s Power Generation business.
Lux Research published a report last year which estimated the size of the water infrastructure repair (drinking and waste water) market at $17 billion a year worldwide. Cities lose large amounts of water through leaks and inefficiencies. Water companies call this  “non-revenue water”, since they treat the water but cannot charge for it since it does not reach customers.
“A city such as London loses 30 percent of its water; over 600 million cubic meters, ” said  TaKaDu’s VP of Marketing Guy Horowitz. “A cubic meter can cost anywhere from a $0.20 to a few dollars, so we’re talking about hundreds of thousands of dollars, if not millions, per day. Indirect losses such as road and property damage, traffic interruptions, paying detection crews and repair crews, and in many cases regulatory fines, amount to hundreds of millions per year in large cities. ” The situation is much worse in the developing world where water that could supply 200 million people is lost every year.
TaKaDu offers water infrastructure monitoring as a service. The company’s software links to existing sensors in the water network like flow meters, quality sensors and pressure meters. Geographical information system (GIS) data, maintenance records, access control records and seasonal fluctuation data is also used. All this data is crunched to detect problems like leaks,  inefficiencies and faults in the water network.  Thames Water, which runs London’s water system, detects leaks up to 9 days earlier with TaKaDu than with its previous system.
TaKaDu uses various data mining techniques to identify problems. One example algorithm is cross-site correlation that matches up neighborhoods which display common water consumption behavior over time.  ”Assume your neighborhood’s consumption goes up by 10 percent while the ‘similar’ neighborhood does not, ” explains Horowitz. “That may indicate a problem, while if both increase it may indicate a warm day, or some other benign explanation.”
In general, the water management business makes minimal usage of IT and is grappling with outdated and decaying infrastructure. Leaks were often only detected manually or by comparing current system behavior to historical behavior, such as last year’s pipe pressure compared to this year. Having an operator inspect a pipe manually inch by inch is still a common practice to track down a local leak, as is replacing an entire pipe. Even small water companies have hundreds of miles of piping. A town of 50,000 typically has more than 200 miles of wastewater pipe alone.
TaKaDu was founded in 2009, has 25 employees and  is based in Yahud, Israel. The new $6 million investment was led by ABB Technology Ventures, ABB’s venture capital arm and also included existing TaKaDu investors Emerald Technology Ventures, Gemini Israel Funds and Giza Venture Capital. TaKaDu raised $3.5 million in 2009.

Red Hat and 10Gen partner to power development of open source data platforms

yamandiuka

Two of the biggest names in the open source world announced a partnership yesterday. Red Hat, the North Carolina based provider of open source software for the enterprise and 10Gen, the New York based company behind the increasingly popular NoSQL database MongoDB, are partnering to help developersdeliver on the promise of big data and the cloud.
While both companies focus on open-source software, they are no slouch in the business department.Red Hat recently hit a big milestone, becoming the first company based around Linux to hit a billion dollars in annual revenue. Last September 10Gen raised a fresh $20 million in venture funding and has been expanding its Manhattan office and hiring like mad.
“Enterprise customers are looking for robust platforms to develop and deploy their next-generation of applications,” said Max Schireson, president of 10gen. “As a leader in the NoSQL community, MongoDB is well-matched to the scalability and agility that developers demand, and through collaboration with Red Hat, expect to create the leading open source enterprise solution for companies deploying next-generation applications.”
The companies hope that clients will become joint customers who can benefit from 10gen’s leading NoSQL database and the performance, stability, and security of the Red Hat platforms, enhancing investments in both firms.
“Web and enterprise developers need solutions that allow them to rapidly deploy applications that deal with large amounts of data in flexible public or private cloud environments,” said Scott Crenshaw, vice president and general manager of the cloud business unit at Red Hat. “Combining Red Hat’s technology stack with 10gen’s MongoDB NoSQL database will help developers to deliver on the promise of big data and cloud technologies.”